Inter-fleet resource sharing using bidding mechanism

Abstract

A system and method for resource sharing between fleets using a bidding mechanism is presented. The fleet scheduler generates a bid proposal for one or more unallocated tasks associated with the first fleet, where the bid proposal is associated with a second fleet, the second fleet being associated with an excess of resources. The online platform receives a bid proposal for the one or more unallocated tasks, and uses the bid proposal and determines one or more winning bids based on the auction scheme. The one or more winning bids include allocating the one or more unallocated tasks from a first fleet to a second fleet with excess resources.

Description

[0001] introduce

[0002] Vehicles are rapidly integrating an ever-increasing number of technology components into their systems. Specialized microcontrollers, technologies, and sensors can be used for many different applications in vehicles. Automotive microcontrollers and sensors can be used to enhance the automation structure that provides customers with state-of-the-art experiences and services, such as in tasks such as body control, camera vision, information display, safety, autonomous control, etc. Additionally, fleet owners can use such technologies to assist in their operations.

[0003] Additionally, some fleet operators may have underutilized resources in terms of vehicles and drivers. Other fleet operators may have overutilized resources in terms of too many tasks. Optimal utilization of vehicles and drivers is critical for fleets to operate efficiently. Utilization of resources may not be consistent and may fluctuate based on demand. Sharing resources between fleets in geographical proximity may help compensate for this fluctuation. Summary of the invention

[0004] A system and method for sharing resources among fleets using a bidding mechanism is disclosed. Real-time data from vehicles can be used as the basis for machine learning price estimates for tasks, where underutilized fleets can predict the cost of completing tasks based on vehicle data such as body type, towing capacity, range, etc.

[0005] Thus, a system for sharing resources among fleets using a bidding mechanism may include a first fleet, wherein the first fleet may be associated with one or more unallocated tasks. In addition, a second fleet may be associated with excess resources. The system may also include a fleet scheduler for the second fleet, and the fleet scheduler may generate a bid proposal for the one or more unallocated tasks. The system may include an online platform that may receive a bid proposal for the one or more unallocated tasks from the fleet scheduler, and based on an auction scheme, one or more winning bids may be determined using the bid proposals, wherein the one or more winning bids may include allocating the one or more unallocated tasks from the first fleet to a second fleet having excess resources.

[0006] Another aspect of the disclosure may include where the bid proposal is based on price estimates for the one or more unassigned tasks for the first fleet.

[0007] Another aspect of the disclosure may include wherein the price estimate is based on vehicle data, driver data, and task data associated with a second fleet of vehicles.

[0008] Another aspect of the present disclosure may include wherein the vehicle data includes body type, towing capacity, range, and distance from a starting location.

[0009] Another aspect of the disclosure may include where the fleet scheduler for the second fleet is further configured to determine an optimal schedule for the second fleet.

[0010] Another aspect of the present disclosure may include wherein the fleet dispatcher is associated with a pool of drivers and vehicles.

[0011] Another aspect of the present disclosure may include where a pool of drivers and vehicles is associated with data that may include driver availability, driver scores and skill sets, vehicle availability, and an initial request price for each driver and vehicle use.

[0012] Another aspect of the present disclosure may include where a fleet scheduler is associated with a task pool.

[0013] Another aspect of the disclosure may include where the pool of tasks is associated with data that may include one or more task characteristics and an initial suggested payment amount for completing the tasks.

[0014] Another aspect of the present disclosure may include where a fleet scheduler may be optimized for task scheduling by bidding on drivers and vehicles.

[0015] Another aspect of the present disclosure may include where a fleet dispatcher is optimized for driver dispatch by bidding on tasks.

[0016] Another aspect of the present disclosure may include wherein the online platform includes a fleet coordinator.

[0017] Another aspect of the disclosure may include a simulation system that may simulate scheduling based on one or more bidding actions.

[0018] Another aspect of the present disclosure may include a fleet manager including budget constraints and weights for expenditure and scheduling optimization.

[0019] Another aspect of the present disclosure may include a method for sharing resources among fleets using a bidding mechanism, the method may include: generating, by a fleet scheduler, a bidding proposal for one or more unassigned tasks associated with a first fleet, wherein the bidding proposal may be associated with a second fleet, the second fleet being associated with excess resources. The method may also include receiving, by an online platform, the bidding proposal for the one or more unassigned tasks, and then determining, by the online platform, one or more winning bids using the bidding proposal and based on an auction scheme, wherein the one or more winning bids may include allocating the one or more unassigned tasks from the first fleet to a second fleet having excess resources.

[0020] Another aspect of the method may include determining, by the fleet dispatcher, an optimal dispatch for the second fleet.

[0021] Another aspect of the method may include determining the best schedule based on optimizing the task schedule by bidding on drivers and vehicles.

[0022] Another aspect of the method may include where determining the optimal schedule is based on optimizing driver schedules by bidding on tasks.

[0023] Another aspect of the method may include where the auction scheme is based on a Vickery-Clarke-Groves mechanism or a Dutch auction mechanism.

[0024] Another aspect of the present disclosure may include a system for sharing resources among fleets using a bidding mechanism, the system may include a first fleet, wherein the first fleet is associated with one or more unassigned tasks. The system may also include a second fleet, wherein the second fleet is associated with excess resources. The system may also include a fleet scheduler for the second fleet, the fleet scheduler may generate a bid proposal for the one or more unassigned tasks. The system may also include an online platform, the online platform may receive bid proposals for the one or more unassigned tasks from the fleet scheduler, and then the online platform may determine one or more winning bids based on an auction scheme using the bid proposals, wherein the one or more winning bids include allocating the one or more unassigned tasks from the first fleet to a second fleet with excess resources. The system may also include a simulation system configured to simulate scheduling based on one or more bidding actions and a fleet manager that may include budget constraints and weights for expenditure and scheduling optimization. The system may include wherein the bid proposal is based on a price estimate for the one or more unassigned tasks of the first fleet. The system may include wherein the price estimate is based on vehicle data, driver data, and task data associated with the second fleet. The system may include wherein the vehicle data includes body type, towing capacity, range, and distance from a starting location. The system may include where a fleet scheduler for a second fleet may determine an optimal schedule for the second fleet. The system may include where the fleet scheduler is associated with a pool of drivers and vehicles. The system may include where the pool of drivers and vehicles is associated with data including driver availability, driver scores and skill sets, vehicle availability, and an initial requested price for each driver and vehicle use. The system may include where the fleet scheduler is associated with a pool of tasks. The system may include where the task pool is associated with data including one or more task characteristics and an initial suggested payment amount for completing the task. The system may include where the fleet scheduler is optimized for task scheduling by bidding on drivers and vehicles, or where the fleet scheduler is optimized for driver scheduling by bidding on tasks. The system may include where the online platform includes a fleet coordinator.

[0025] A first aspect of the present disclosure provides a system for sharing resources among fleets using a bidding mechanism, comprising:

[0026] a first fleet, wherein the first fleet is associated with one or more unassigned tasks;

[0027] a second fleet, wherein the second fleet is associated with the excess resource;

[0028] a fleet scheduler for a second fleet of vehicles configured to generate bid proposals for the one or more unassigned tasks;

[0029] an online platform configured to receive bid proposals for the one or more unassigned tasks from the fleet dispatcher; and

[0030] The online platform is also configured to determine one or more winning bids using the bid proposals based on an auction scheme, wherein the one or more winning bids include allocating the one or more unassigned tasks from the first fleet to a second fleet having excess resources.

[0031] According to the system of the first aspect of the present disclosure, the bid proposal is based on price estimates of the one or more unassigned tasks of the first fleet.

[0032] The system according to the first aspect of the present disclosure, wherein the price estimate is based on vehicle data, driver data, and task data associated with the second fleet.

[0033] According to the system described in the first aspect of the present disclosure, the vehicle data includes body type, traction capacity, cruising range and distance from a starting location.

[0034] According to the system described in the first aspect of the present disclosure, the fleet dispatcher for the second fleet is further configured to determine an optimal dispatch for the second fleet.

[0035] According to the system of the first aspect of the present disclosure, the fleet dispatcher is associated with a pool of drivers and vehicles.

[0036] The system according to the first aspect of the present disclosure, wherein the driver and vehicle pool is associated with data including: driver availability, driver score and skill set, vehicle availability, and initial request price for each driver and vehicle use.

[0037] According to the system described in the first aspect of the present disclosure, the fleet scheduler is associated with the task pool.

[0038] According to the system described in the first aspect of the present disclosure, the task pool is associated with data including: one or more task characteristics and an initially suggested payment amount for completing the task.

[0039] According to the system described in the first aspect of the present disclosure, the fleet dispatcher is optimized for task scheduling by bidding on drivers and vehicles.

[0040] According to the system described in the first aspect of the present disclosure, the fleet dispatcher is optimized for driver scheduling by bidding on tasks.

[0041] According to the system described in the first aspect of the present disclosure, the online platform includes a fleet coordinator.

[0042] The system according to the first aspect of the present disclosure further includes a simulation system configured to simulate scheduling based on one or more bidding actions.

[0043] The system according to the first aspect of the present disclosure further includes a fleet manager, wherein the fleet manager includes budget constraints and weights for expenditure and scheduling optimization.

[0044] A second aspect of the present disclosure provides a method for sharing resources among fleets using a bidding mechanism:

[0045] generating, by a fleet scheduler, a bid proposal for one or more unassigned tasks associated with a first fleet, wherein the bid proposal is associated with a second fleet, the second fleet being associated with excess resources;

[0046] receiving, by the online platform, bid proposals for the one or more unassigned tasks; and

[0047] The online platform uses the bid proposals and determines one or more winning bids based on an auction scheme, wherein the one or more winning bids include allocating the one or more unassigned tasks from the first fleet to a second fleet having excess resources.

[0048] The method according to the second aspect of the present disclosure further includes determining, by the fleet scheduler, an optimal schedule for the second fleet.

[0049] The method according to the second aspect of the present disclosure, wherein determining the best schedule is based on optimizing the task schedule by bidding on drivers and vehicles.

[0050] The method according to the second aspect of the present disclosure, wherein determining the optimal schedule is based on optimizing driver scheduling by bidding on tasks.

[0051] According to the method of the second aspect of the present disclosure, the auction scheme is based on a Vickery-Clarke-Groves mechanism or a Dutch auction mechanism.

[0052] A third aspect of the present disclosure provides a system for sharing resources among fleets using a bidding mechanism, comprising:

[0053] a first fleet, wherein the first fleet is associated with one or more unassigned tasks;

[0054] a second fleet, wherein the second fleet is associated with the excess resource;

[0055] a fleet scheduler for a second fleet of vehicles configured to generate bid proposals for the one or more unassigned tasks;

[0056] an online platform configured to receive bid proposals for the one or more unassigned tasks from the fleet dispatcher;

[0057] The online platform is further configured to determine one or more winning bids using the bid proposals based on an auction scheme, wherein the one or more winning bids include assigning the one or more unassigned tasks from the first fleet to a second fleet having excess resources;

[0058] a simulation system configured to simulate scheduling based on the one or more bidding actions; and

[0059] Fleet manager, including budget constraints and weights for expenditure and scheduling optimization;

[0060] wherein the bid proposal is based on price estimates for the one or more unassigned tasks of the first fleet,

[0061] wherein the price estimate is based on vehicle data, driver data, and mission data associated with the second fleet,

[0062] The vehicle data includes vehicle type, towing capacity, range and distance from the starting location.

[0063] wherein the fleet scheduler for the second fleet is further configured to determine an optimal schedule for the second fleet,

[0064] wherein the fleet dispatcher is associated with a pool of drivers and vehicles,

[0065] wherein the pool of drivers and vehicles is associated with data including driver availability, driver scores and skill sets, vehicle availability, and an initial request price for each driver and vehicle use,

[0066] The fleet scheduler is associated with the task pool.

[0067] The task pool is associated with data including: one or more task characteristics and an initial suggested payment amount for completing the task,

[0068] wherein the fleet scheduler is optimized for task scheduling by bidding on drivers and vehicles, or wherein the fleet scheduler is optimized for driver scheduling by bidding on tasks, and

[0069] The online platform includes a fleet coordinator.

[0070] The above-mentioned features and advantages of the present disclosure and other features and incidental advantages will be apparent from the following detailed description of illustrative examples and modes for carrying out the present disclosure when understood in conjunction with the accompanying drawings and the appended claims. In addition, the present disclosure expressly includes combinations and sub-combinations of elements and features presented above and below. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0072] Figure 1 is a diagram of a fleet telemetry system in accordance with the present disclosure.

[0073] Figure 2 is a flow chart of a system for sharing resources among fleets of vehicles for optimizing task scheduling by bidding on drivers and vehicles according to the present disclosure.

[0074] Figure 3 is a flow chart of a system for sharing resources among fleets of vehicles for optimizing driver scheduling by bidding on tasks according to the present disclosure.

[0075] Figure 4 A fleet resource auction system according to the present disclosure.

[0076] Figure 5 is a flow chart of a method for sharing resources among fleets using a bidding mechanism according to the present disclosure.

[0077] The accompanying drawings are not necessarily drawn to scale and may present somewhat simplified representations of various preferred features of the present disclosure disclosed herein (including, for example, specific dimensions, orientations, locations, and shapes). The details associated with these features will be determined in part by the specific intended application and use environment. DETAILED DESCRIPTION

[0078] The present disclosure may allow for embodiments in many different forms. Representative examples of the present disclosure are shown in the accompanying drawings and described in detail herein as non-limiting examples of the disclosed principles. For this reason, elements and limitations described in the abstract, introduction, summary, and detailed description but not explicitly stated in the claims should not be incorporated into the claims individually or collectively by implication, inference, or otherwise.

[0079] For the purposes of this description, unless expressly disclaimed, the use of the singular includes the plural and vice versa, the terms "and" and "or" shall be conjunctive and disjunctive, and the words "including", "comprising", "containing", "having" and the like shall mean "including but not limited to". In addition, approximate words such as "approximately", "almost", "substantially", "generally", "approximately" and the like may be used herein in the sense of "at, close to or substantially at" or "within 0-5%" or "within an acceptable manufacturing tolerance" or logical combinations thereof. As used herein, a component "configured to" perform a specified function is capable of performing the specified function without change, rather than merely having the potential to perform the specified function after further modification. In other words, when expressly configured to perform a specified function, the hardware described is specifically selected, created, implemented, utilized, programmed and / or designed for the purpose of performing the specified function.

[0080] Referring to the drawings, the left-most digit(s) of a reference number identifies the drawing in which the reference number first appears (e.g., reference number "310" indicates that the element so numbered appears in the drawing). Figure 3 Additionally, elements having the same reference numeral followed by a different letter or other distinguishing mark (e.g., an apostrophe) indicate elements that may be identical in structure, operation, or form but may be identified as being at different locations in space or appearing repeatedly at different points in time (e.g., reference numerals "110a" and "110b" may indicate two different input devices that may be identical in function but may be located at different points in the simulated venue).

[0081] Vehicles have become computationally advanced and equipped with multiple microcontrollers, sensors, and control systems, including, for example, autonomous vehicles and advanced driver assistance systems (AV / ADAS), such as adaptive cruise control, automatic parking, automatic brake hold, automatic braking, evasive steering assist, lane keeping assist, adaptive headlights, reversing assist, blind spot detection, cross traffic alert, local hazard alert, and rear automatic braking may depend on information obtained from cameras and sensors on the vehicle. Such systems can also provide a wealth of information about the vehicle, including, for example, location, automatic assist sensors, occupancy data, motion sensors, and so on. These systems can also provide detailed information about the operation of the vehicle that may not be realized by its occupants. This information can be compiled into an online platform through the cloud for further analysis. For example, a fleet owner can collect vehicle data for a specific driver to determine the driving style and habits of that driver, thereby determining a driver profile.

[0082] Fleet operators may often face problems with resource and task assignment. Too many unassigned tasks may mean over-utilization of fleet resources. Too few unassigned tasks may mean under-utilization of the fleet. An online platform may be used to balance tasks and resources between fleet managers using an online platform with an auction system. Such a system may then allow some or all unassigned tasks to be assigned from one fleet to another fleet with excess resources. Such auctions of tasks and resources may also be based on an optimized schedule based on vehicle status data that may be collected using a fleet telemetry system.

[0083] Figure 1 1 is a diagram of a fleet telemetry system 100 according to an embodiment of the present disclosure. The system 100 may include a fleet of vehicles including vehicles of different types and styles, such as tractor trailers 110, trucks 115, or two-door or four-door sedans 120 or other types of vehicles. The vehicles may communicate with an online platform 140 including a storage device 150 via the Internet (e.g., as shown in cloud 130). The system 100 may be configured so that the vehicle (e.g., tractor trailer 110, truck 115, or sedan 120) may collect information about the driver's driving performance and vehicle data through internal sensors, controllers, and processors. The driving information may include trip data, which may be analyzed for attributes such as aggressiveness, fuel / energy efficiency, and associated maintenance costs based on the type of vehicle driven and the type of road the vehicle is traveling on. The vehicle information may include data such as body type, towing capacity, number of passenger seats, drive system type, electric vehicle status, location, etc. This information may then be sent to a server such as an online platform 140 and a storage device 150 via the cloud 130. Additionally, the trip history may be stored in the storage device 150 for further analysis.

[0084] Figure 2 A flow chart of a system 200 for optimizing task scheduling by bidding on drivers and vehicles according to an embodiment of the present disclosure is illustrated. The system 200 may include a fleet coordinator 250, a geographic grid 210, and a fleet manager 230. The fleet coordinator 250 may also include a fleet 215, a scheduler 220, a driver and vehicle pool 225, a simulator 235, and a bidding system 240.

[0085] The system 200 can be broken down or segmented according to a geographic grid 210. The grid can be defined in different ways, such as by map coordinates, by regions such as zip codes, or other metrics. The goals of the geographic grid 210 include mapping the proximity of resources (such as from the driver and vehicle pool 225) to the fleet, and the proximity can be analyzed at each geographic unit (e.g., geographic unit 1, geographic unit 2, up to geographic unit N).

[0086] As a single entity, the fleet coordinator 250 can use the scheduler 220 to optimize the scheduling of all fleets (e.g., fleet 1, fleet 2, up to fleet N) in the fleet 215, and the scheduler 220 can determine the best schedule for each fleet within the limit of X days, assuming that X can be a number of days or a portion of a day. Assuming each optimal schedule, such as the best schedule for fleet 1, the best schedule for fleet 2, and up to the best schedule for fleet N, the scheduler 220 can then automatically assign available drivers and vehicles to the driver and vehicle pool 225. The driver and vehicle pool 225 can also be referred to as a market, which allows other fleets to bid for unallocated resources. The driver and vehicle pool 225 is based on the scheduler 220 can determine the available drivers and vehicles for each time slot across all relevant fleets. Information about drivers and their skill sets can also be matched manually or automatically with the needs of interested bidders for available resources.

[0087] Utilizing the output of the scheduler 220 and the driver and vehicle pool 225, the bidding system 240 can then allow fleets with excess or unassigned tasks to bid for available driver and vehicle resources. Fleets can compete for the use of other fleet drivers and vehicles via the bidding system 240. In addition, the bidding process can be automated by the fleet agent attempting to maximize its fleet schedule under budget constraints. In addition, the bidding process can be done manually by a fleet manager. When the bidding process begins, the scheduler 220 can further optimize the best schedule for each fleet. In addition, because the bidding system 240 continues to receive bids, the optimization process can be iterative.

[0088] The simulator 235 can then simulate dispatching based on using fleet drivers and vehicles through bidding actions. In addition, based on the machine learning price estimator, the fleet manager can be updated about potential revenue if a resource is recommended, or if the fleet manager is a bidder, the cost and dispatch benefits can be analyzed.

[0089] The machine learning price estimator can take as input data from each fleet vehicle, which can be transmitted in real time by the fleet vehicles. Such inputs may include vehicle data such as: body type, towing capacity, number of passenger seats, drive system type, vehicle current range, electric vehicle charging capacity (if applicable), and distance from the starting location of the task. Other inputs may include driver information, such as driver score or skill set. Task information can also be an input to the machine learning price estimator and include data such as the start time and location of the task, task duration, distance to be traveled, and estimates of fuel or energy to be used. The machine learning regression model can then use these inputs to make continuous variable predictions using regression analysis, enhancement methods, or neural networks. The output of the machine learning regression model can be a predicted task completion cost, which can be used as an estimate for initial task price calibration that can be used in a bid price proposal.

[0090] The fleet manager 230 , based on interactive communications with the simulator 235 , may impose budget constraints or weights on the expenditure and scheduling optimization used by the simulator 235 .

[0091] Figure 3 A flow chart of a system 300 for optimizing driver scheduling by bidding on tasks according to an embodiment of the present disclosure is illustrated. The system 300 may include a fleet coordinator 350, a geographic grid 210, and a fleet manager 230. The fleet coordinator 350 may also include a fleet 215, a scheduler 220, a task pool 325, a simulator 335, and a task bidding system 340.

[0092] System 300 may be decomposed or segmented according to geographic grid 210 as discussed in system 200. As a single entity, fleet coordinator 350 may optimize the scheduling of all fleets (e.g., fleet 1, fleet 2, up to fleet N) in fleet 215 using scheduler 220, which may determine the best schedule for each fleet within a limit of X days, assuming that X may be a number of days or a portion of a day.

[0093] When optimizing driver scheduling by bidding on tasks, based on scheduler 220, task pool 325 may include tasks that are at risk of not being completed on time or have low priority due to driver or vehicle unavailability. As with system 200, system 300 may use a machine learning-based price estimator to determine an initial suggested payment amount for completing a task based on task requirements (e.g., desired destination, desired vehicle type, time frame, etc.).

[0094] Using the output of the scheduler 220 and the task pool 325, the task bidding system 340 can then allow fleets with excess resources to bid to perform other fleet tasks. Fleets can compete for other fleet tasks via the task bidding system 340. In addition, the bidding process can be completed automatically by the fleet agent trying to maximize driver and vehicle utilization. In addition, the bidding process can be completed manually by the fleet manager competing for tasks. When the bidding process begins, the scheduler 220 can further optimize the previous best schedule for each fleet. In addition, because the bidding system 340 updates the schedule based on the tasks won during the bidding process, the optimization process can be iterative.

[0095] The simulator 335 can then simulate scheduling based on the tasks obtained through the bidding actions. In addition, as discussed in the system 200, the system 300 with the simulator 335 can update the fleet manager based on the machine learning price estimator regarding the potential revenue of bidding a particular task at a particular price.

[0096] The fleet manager 230 may impose budget constraints or weights on the expenditure and scheduling optimization used by the simulator 335 based on interactive communications with the simulator 335 .

[0097] The task bidding system 340 can generate a combinatorial auction as a mechanism for bidding on multiple tasks simultaneously. Phase 1 of the mechanism can include determining one or more resource availability windows. Thus, phase 1 can include determining an availability window for each resource in the fleet, which can include an initial location and start time and an estimated end location and end time.

[0098] Phase 2 may include checking the task pool, which may include obtaining information about the task requirements and durations for each task in the task pool 325. For example, for task #1, the source location and start time, duration, and target location and end time. The same will apply to task #2, task #3, etc.

[0099] Phase 3 may include simulated scheduling of tasks utilizing the pool, which may include possible groupings of tasks during the availability window from Phase 1 based on price and cost estimators (such as the machine learning regression model discussed previously), assuming available resources.

[0100] Figure 4 A fleet resource auction system 400 according to an embodiment of the present disclosure is illustrated. The fleet resource auction system 400 may include an auction system 410, an overutilized fleet 420, and an underutilized fleet 430. The auction system 410 may include a bid repository 412, an auction scheme 414, and a global coordinator 416.

[0101] Overutilized fleet 420 may include multiple fleets, for example, as shown by fleet 1, fleet 2, and so on to fleet N. Overutilized fleet 420 may be associated with one or more unassigned tasks that have not been completed due to lack of resources, low priority, etc. Each unassigned task may also be associated with a start and end location, a start time, an end time, and an associated duration.

[0102] The underutilized fleet 430 may include multiple fleets, for example, as shown by Fleet 1, Fleet 2, up to Fleet N. The underutilized fleet 430 may be associated with one or more available resources as indicated by Scheduler 1, Scheduler 2, up to Scheduler N, each scheduler being associated with its corresponding fleet and indicating the availability of each fleet's resources.

[0103] The auction system 410 may accept input into its bid repository 412 of one or more unassigned tasks from an overutilized fleet 420 and bids from an underutilized fleet 430. The bids from the underutilized fleet 430 may be based on driver and vehicle availability, time, and location as indicated by each dispatcher. As previously discussed, each bid from a dispatcher may be based on a price estimate using vehicle data, driver data, and task data associated with a fleet resource.

[0104] Once the bid repository 412 receives the unassigned tasks from the overutilized fleet 420 and the initial bids from the schedulers of the underutilized fleet 430, the auction scheme 414 can determine one or more winning bids using methods such as a Vickrey-Clark-Groves mechanism or a Dutch auction. Then, based on the winning bids from the auction scheme 414, the global coordinator 416 can dispatch the unassigned tasks from the overutilized fleet 420 and assign them to the corresponding winning fleets within the underutilized fleet 430. The global coordinator 416 can also make scheduling modification recommendations between fleets within the underutilized fleet 430. For example, the global coordinator 416 can recommend that a first task be assigned to driver B and another task be assigned to driver A so as to enable driver B to complete a different third task. Such recommendations can be made using a potential algorithmic solution based on a job shop scheduling solver, such as Johnson's algorithm.

[0105] Figure 5An exemplary embodiment of a method 500 for sharing resources among fleets using a bidding mechanism according to an embodiment of the present disclosure is shown. The method 500 begins at step 505, where a fleet scheduler generates a bid proposal for one or more unassigned tasks associated with a first fleet, wherein the bid proposal is associated with a second fleet, and the second fleet is associated with excess resources.

[0106] like Figure 3 As discussed in, fleet coordinator 350 can use scheduler 220 to optimize the scheduling of all fleets (e.g., fleet 1, fleet 2, up to fleet N) in fleet 215, and scheduler 220 can determine the best schedule for each fleet within a limit of X days, assuming that X can be a number of days or a portion of a day. System 300 can also include a process for optimizing driver scheduling by bidding on tasks, based on scheduler 220, task pool 325 can include tasks that are at risk of not being completed on time or have low priority due to driver or vehicle unavailability. As with system 200, system 300 can use a machine learning-based price estimator to determine an initial suggested payment amount for completing a task based on task requirements (e.g., desired destination, desired vehicle type, time frame, etc.).

[0107] At step 510, the online platform may receive bid proposals for the one or more unassigned tasks. Figure 4 As discussed in, the auction system 410 can accept one or more unassigned tasks from the overutilized fleet 420 and bids from the underutilized fleet 430 into its bid repository 412. The bids from the underutilized fleet 430 can be based on driver and vehicle availability, time, and location indicated by each dispatcher. As discussed, each bid from the dispatcher can be based on a price estimate using vehicle data, driver data, and task data associated with the fleet resources.

[0108] At step 515, the online platform may use the bid proposals and determine one or more winning bids based on an auction scheme, wherein the one or more winning bids include allocating one or more unassigned tasks from the first fleet to a second fleet with excess resources. Figure 4As discussed in , once the bids are received, the repository 412 accepts the unassigned tasks from the overutilized fleet 420 and the initial bids from the schedulers of the underutilized fleet 430, and then the auction scheme 414 can use a method such as a Vickrey-Clark-Groves mechanism or a Dutch auction to determine one or more winning bids. Then, based on the winning bids from the auction scheme 414, the global coordinator 416 can dispatch the unassigned tasks from the overutilized fleet 420 and assign them to the corresponding winning fleets within the underutilized fleet 430.

[0109] Method 500 may then end.

[0110] The Description and Abstract sections may set forth one or more embodiments of the present disclosure as contemplated by the inventor(s), and thus are not intended to limit the present disclosure and the appended claims.

[0111] Embodiments of the present disclosure have been described above with the aid of functional building blocks that illustrate implementations of specific functions and their relationships. For ease of description, the boundaries of these functional building blocks have been arbitrarily defined herein. Alternative boundaries may be defined as long as the specified functions and their relationships can be properly performed.

[0112] The foregoing description of the specific embodiments will fully reveal the general nature of the present disclosure, so that others can easily modify and / or adapt the various applications of such specific embodiments by applying knowledge within the technical scope of the art without departing from the general concept of the present disclosure, without excessive experimentation. Therefore, based on the teachings and guidance presented herein, such adaptations and modifications are intended to be within the meaning and scope of the equivalents of the disclosed embodiments. It should be understood that the wording or terminology herein is for the purpose of description rather than limitation, so that the terms or wording of this specification will be interpreted by those skilled in the art according to teaching and guidance.

[0113] The breadth and scope of the present disclosure should not be limited by the above-described exemplary embodiments.

[0114] Exemplary embodiments of the present disclosure have been presented. The present disclosure is not limited to these examples. These examples are presented herein for illustrative and non-limiting purposes. Based on the teachings contained herein, alternatives (including equivalents, extensions, variations, deviations, etc. of those described herein) will be apparent to those skilled in the relevant art(s). These alternatives fall within the scope and spirit of the present disclosure.

Claims

1. A system for sharing resources among fleets using a bidding mechanism: a first fleet, wherein the first fleet is associated with one or more unassigned tasks; a second fleet, wherein the second fleet is associated with the excess resource; a fleet scheduler for a second fleet of vehicles configured to generate bid proposals for the one or more unassigned tasks; an online platform configured to receive bid proposals for the one or more unassigned tasks from the fleet dispatcher; as well as The online platform is also configured to determine one or more winning bids using the bid proposals based on an auction scheme, wherein the one or more winning bids include allocating the one or more unassigned tasks from the first fleet to a second fleet having excess resources.

2. The system of claim 1, wherein the bid proposal is based on price estimates for the one or more unassigned tasks of the first fleet. 3 . The system of claim 2 , wherein the price estimate is based on vehicle data, driver data, and mission data associated with the second fleet.

4. The system of claim 3, wherein the vehicle data includes body type, towing capacity, range, and distance from a starting location. 5 . The system of claim 1 , wherein the fleet scheduler for a second fleet is further configured to determine an optimal schedule for the second fleet.

6. The system of claim 1, wherein the fleet dispatcher is associated with a pool of drivers and vehicles.

7. The system of claim 6, wherein the pool of drivers and vehicles is associated with data including driver availability, driver scores and skill sets, vehicle availability, and an initial request price for each driver and vehicle use.

8. The system of claim 1, wherein the fleet scheduler is associated with the task pool.

9. The system of claim 8, wherein the task pool is associated with data including: one or more task characteristics and an initial suggested payment amount for completing the task.

10. The system of claim 1, wherein the fleet scheduler is optimized for task scheduling by bidding on drivers and vehicles.

Images